Impact of the Rise of Generative AI on Developing Countries | IGF 2023 Town Hall #29

Audience

The discussion revolves around the impact of technology, particularly AI, on freedom and democracy. There is a neutral sentiment overall, with a focus on whether AI will lead to more freedom and democracy or have the opposite effect. One argument is that authoritarian rulers might use technology to establish more control rather than promote freedom. This raises concerns about the potential misuse of AI in authoritarian regimes.

Moving on to the potential application of generative AI in developing countries, the sentiment remains neutral. It is recognised that generative AI has the potential to benefit these nations, although specific evidence or supporting facts are not provided. Nonetheless, there is an interest in exploring the application of generative AI in developing countries, highlighting a desire to leverage technology for their development.

Another aspect discussed is the need to redefine the term ‘developing countries.’ This argument emphasises the existence of highly functional digital societies in Estonia, Finland, Norway, and the Netherlands. These societies serve as examples of how advancements in technology can lead to progress and development. The recommendation is to learn from these societies and implement their successes in other parts of the world. Young people are seen as crucial in this process, as they can observe and learn from these digital societies, then bring back knowledge to design and restructure their own societies. The high youth population in regions like India, the continent, and the Middle East and North Africa (MENA) further amplifies the importance of involving the young generation in shaping the future.

The impact of AI on human creativity and its contribution to human resources is considered in a neutral sentiment, without specific arguments or evidence provided. The broader question of how AI will affect human creativity and its implications for the workforce remains unanswered.

Concerns are raised about the use of native AI in developing countries with limited resources and infrastructure. The sentiment is concerned, with a focus on the potential widening of the technology gap in these nations. The argument questions whether native AI will exacerbate inequalities and further marginalise resource-poor countries.

In conclusion, this analysis highlights various viewpoints on the impact of technology, specifically AI, on freedom, democracy, development, and creativity. While concerns are raised regarding the potential misuse of AI and widening technology gaps, there is still potential for positive outcomes through the application of AI in developing countries. The role of young people and learning from successful digital societies are also emphasised in shaping a better future for societies worldwide.

Atsushi Yamanaka

In a recent analysis, different viewpoints on the topic of Artificial Intelligence (AI) were discussed. Atsushi Yamanaka, a senior advisor on digital transformations at JICA, shared his belief that AI has both significant potential and notable threats. With 28 years of experience in the field, Yamanaka advises JICA on incorporating technology elements into various projects and supporting digital transformation initiatives.

One area where Yamanaka sees promising potential is the use of generative AI models for local African languages. He highlighted how this application could play a crucial role in promoting digital inclusion in developing economies. By developing AI models for African languages, barriers in digital literacy could be overcome, enabling more people to benefit from technology. Yamanaka’s colleague, who studied AI in Japan and now works at Princeton, is actively working on generative AI models for African languages.

However, the analysis also acknowledged the potential risks associated with the rise of generative AI. It highlighted the concern that this technology could lead to an increase in misinformation, making it increasingly difficult to distinguish between real and fake information. As a result, trust in digital technology may be undermined, presenting challenges for individuals and societies. These issues underscore the importance of responsible development and deployment of AI technologies.

Another argument made in the analysis was the need to establish a consensual framework for AI regulations. The participation of emerging countries was emphasized, as developing nations should play an active role in global discussions on AI regulations. The aim is to avoid creating multiple fragmented models or regulations and instead work towards a unified approach that addresses the concerns and interests of all stakeholders.

Concerns were raised about the potential impact of AI technology on labor. The recent work by the International Labor Organization indicates potential job losses resulting from the introduction of AI. In the United States, for example, the Screenwriters Guild has expressed concerns about AI replacing their jobs, sparking fears of a potential backlash reminiscent of the Luddite Movement of the 19th century. These concerns emphasize the need to consider the potential negative consequences on employment and to ensure that appropriate measures are taken to mitigate any adverse impacts.

Privacy invasion was another aspect discussed in the analysis. The Chinese AI-based scoring system was highlighted as an example of technology that invades privacy. The system reportedly monitors and scores every aspect of citizens’ lives. This raises concerns among privacy advocates and highlights the ethical considerations that need to be taken into account as AI technologies continue to evolve and become more integrated into daily life.

The analysis also touched upon the digital gap between developed and emerging economies. The argument was made that AI technology, particularly new technologies, could actually help reduce this gap. Unlike traditional barriers to communication, there are no interaction barriers in digital technologies, making their adoption in developing countries more feasible. Furthermore, emerging economies might even contribute more to the growth and development of these technologies.

Interestingly, the analysis noted that developing countries have the potential to be at the forefront of innovation in AI technologies. It emphasized that a significant amount of innovation is already emerging from these regions and suggested that they might contribute more to innovation in digital technologies than Western countries. This insight challenges the notion that developing countries will necessarily lag behind in the adoption and advancement of AI technologies.

In conclusion, the analysis delved into various aspects of AI and provided different perspectives on its potential, risks, and implications. It emphasized the need for responsible development, consensual regulatory frameworks, and the active participation of emerging countries in shaping AI technologies. While acknowledging the threats and challenges associated with AI, the analysis also highlighted the opportunities for promoting digital inclusion and reducing the digital gap. Ultimately, it asserted that each society should have the agency to manage its own governance in line with its specific needs and circumstances.

Robert Ford Nkusi

Robert Ford Nkusi is a prominent figure in the field of software testing qualifications in Rwanda. He is currently leading a software testing qualifications team and has made significant contributions to the Rwanda Software Testing Qualifications Board. This demonstrates his expertise and leadership in the industry.

Furthermore, Robert has been involved in the design of the implementation plan for the Child Online Protection Policy in Rwanda. Working under the United Nations, he played a crucial role in developing a comprehensive plan to safeguard children online. This highlights his commitment to promoting child safety and creating a secure digital environment for young users.

In addition to his work in software testing and child protection, Robert has also contributed to the regional framework for the one network area in the East African region. By circumventing data roaming costs, this initiative has greatly benefited individuals and businesses in the area. Robert was actively involved in setting up the one network area and played a vital role in the successful proof of concept testing.

Notably, Robert has led efforts in managing cross-border mobile financial services, which have become increasingly popular in the East African community. By facilitating convenient and secure transactions across borders, these services have contributed to economic growth and poverty reduction. Robert’s involvement in this area demonstrates his expertise in the intersection of finance and technology.

Currently, Robert is engaged with JICA (Japan International Cooperation Agency) in implementing an ICT industry promotion project in Uganda. The four-year project aims to build capacity in the ICT industry and foster innovation and infrastructure development. By leveraging international partnerships, Robert is actively working towards advancing the ICT sector in Uganda.

The potential of generative AI in predicting and mitigating harmful online content was discussed. Robert highlighted how AI can aid in keeping children safe online through the issue of Child Online Protection. However, caution is required with the implementation of generative AI, as its accurate responses can make users less questioning, potentially leading to unforeseen negatives. It is crucial to strike a balance between the benefits and risks of this technology.

Moreover, African countries have shown exemplary progress in implementing and regulating AI technologies, challenging the traditional divide between developed and developing economies. By successfully adopting and regulating AI, these countries have demonstrated their capability in technological advancements.

The debate on long-term leadership and its relationship with authoritarianism was also explored. The definition of democracy and authoritarianism can differ based on the context, and it was argued that long-term leadership is not necessarily synonymous with authoritarianism. This raises important questions about the nature of political leadership and the impact it has on governance.

Furthermore, the potential of generative AI to transform politics was highlighted. The use of AI in predicting political outcomes and shaping political discourse has the ability to revolutionize the political landscape. However, it is important to critically analyze the impact of AI on democratic processes and ensure that it is used responsibly and ethically.

An interesting observation arising from the analysis is the need for collective efforts and shared learning in policy-making for AI technologies. Developing economies, like those in Africa, have successfully implemented technological solutions such as mobile money. It is suggested that developed countries, such as those in the G7, can learn from these successes and collaborate in policy-making to ensure the responsible use of AI technologies.

In conclusion, Robert Ford Nkusi is a leader in software testing qualifications, with notable contributions in the fields of child protection, regional frameworks, cross-border financial services, and ICT industry promotion. The potential and challenges of generative AI were explored, along with the successful implementation of AI technologies in African countries. The complex relationship between long-term leadership and authoritarianism was discussed, and the transformative potential of AI in politics was examined. Overall, these insights shed light on the intersection of technology, governance, and societal progress.

Sarayu Natarajan

The discussion examines the implications of generative AI across various domains, acknowledging its advantages and disadvantages, especially in relation to technology and society. It highlights the study of algorithmic and platform-mediated work, digitization, and digital infrastructure in the context of generative AI, with a focus on effective government-citizen communication.

Data governance is identified as a critical area of focus within generative AI, necessitating exploration of sustainability financing, governance, and digital system replication worldwide. The discussion also raises concerns about the impact of generative AI on the labor market, particularly in developing regions, where workers involved in data annotation and labeling are often overlooked in broader AI conversations.

Furthermore, limitations and biases in data structures restrain the full potential of generative AI, particularly in addressing gender and race representation in the developing world. The potential for generative AI to propagate disinformation and misinformation is also highlighted as a significant concern.

To address these issues and ensure meaningful digital lives and futures, the discussion emphasizes the need for governance and regulation to be considered during AI deployment. Inclusive frameworks of governance and regulation involving global participants are deemed essential to manage the impact of AI across all regions and promote equitable outcomes.

Additionally, the role of generative AI in the creative domain is explored, with the recognition that it can assist in certain types of literature creation. However, it is underlined that the education system and society should continue fostering creativity to avoid over-reliance on AI.

Overall, the analysis delves into the multifaceted implications of generative AI, highlighting the importance of governance, fairness, and ethics. The discussion emphasizes the need for thoughtful and inclusive approaches to harness the potential benefits of generative AI while mitigating its challenges.

Tomoyuki Naito

During the IGF 2023 sessions, Prime Minister Kishida emphasized the importance of generative AI and knowledge sharing for all participants. This highlights the recognition that generative AI has the potential to greatly impact various sectors, and therefore, it should be accessible to everyone.

The discussions at the IGF have brought international experts together, who have acknowledged the threats posed by generative AI. This recognition has sparked thoughts on how to counter these potential threats. The fact that these concerns are widely recognized at an international level shows the serious consideration being given to generative AI.

Tomoyuki Naito, the moderator of the session, specifically emphasized the need to explore the opportunities and threats of generative AI in the context of global south economies. This highlights the importance of understanding how generative AI can impact the economic growth and development of these regions. By recognizing the specific challenges faced by global south economies, tailored strategies can be developed to leverage generative AI for their benefit.

The panel discussion aimed to gather expert opinions on the threats and opportunities presented by generative AI. The first half of the discussion was dedicated to capturing the perspectives and insights of the panelists to ensure a comprehensive understanding of the various viewpoints. The second half of the discussion, on the other hand, was planned to encourage public opinions and comments, fostering a more inclusive and democratic approach to addressing these issues.

Naito’s belief that experts are actively working on addressing the potential threats to privacy and security posed by generative AI is significant. It indicates that these concerns are not being overlooked, and there is a collective effort to develop strategies to mitigate these risks. The fact that many sessions at the IGF have already discussed the potential threats of generative AI further strengthens the notion that this is a widely recognized issue. Moreover, the concerns shared by international experts highlight the seriousness with which these potential threats are being taken.

One noteworthy observation from the discussions is the recognition that countries can proactively utilize new technologies, such as generative AI, for their own economic and social development. This signifies a shift in mindset, where new technologies are seen as opportunities rather than threats. By leveraging these technologies effectively, countries can drive economic growth and social progress.

In conclusion, the IGF 2023 sessions shed light on the importance of generative AI and knowledge sharing for everyone, specifically in the context of global south economies. The discussions recognized the potential threats posed by generative AI and emphasized the need for expert opinions and public engagement. However, there is a collective effort to address these threats and proactively utilize new technologies for economic and social development. Overall, the sessions provided valuable insights and highlighted the significance of inclusive and informed decision-making in the field of generative AI.

Safa Khalid Salih Ali

Generative AI has emerged as a powerful tool with the potential to revolutionize various sectors. The analysis reveals several key benefits and applications of generative AI. Firstly, it can significantly reduce the time spent on data analysis by automating routine tasks. This allows businesses and organizations to derive insights more quickly and efficiently. By eliminating the need for manual data analysis, generative AI enables professionals to focus on improving the quality of their work, ultimately enhancing productivity.

Furthermore, generative AI can play a crucial role in predicting and managing economic crises. By utilising generative AI, experts can develop prediction models that help identify potential crises and take preventive measures accordingly. This is particularly relevant to sectors such as finance and banking, where generative AI can aid in risk assessment and fraud detection. By analysing historical data, generative AI can predict consumer behaviour and help financial institutions make informed decisions to mitigate risks and enhance security.

In the realm of fintech, generative AI has the potential to enhance customer experiences. By providing immediate solutions in emergency situations, generative AI can improve customer satisfaction levels. Additionally, generative AI can democratise financial services by allowing all participants to easily access the services they need, such as virtualized access through chatbots. This fosters financial inclusion and reduces inequalities by ensuring that all citizens have equal opportunities in accessing financial services.

Another significant application of generative AI lies in policy simulation. By simulating the effects of different policies, generative AI can assist policymakers in addressing weaknesses and making informed decisions. Through simulation, potential issues can be identified and resolved before they negatively impact society. For example, the analysis highlights a situation in Sudan where a war could have been preempted if generative AI had been used to simulate the consequences of certain policies.

While the benefits of generative AI are clear, it is crucial to address certain challenges. Developing countries face significant data challenges and lack the necessary knowledge and infrastructure to fully harness the potential of AI. Therefore, it is essential to establish systems that support AI development in these countries. By doing so, they can benefit from the transformative power of generative AI and drive economic growth.

In conclusion, generative AI has immense potential to revolutionize various sectors and bring about significant benefits. Its ability to streamline data analysis, aid in predicting and managing economic crises, enhance customer experiences, simulate policies, and foster financial inclusion makes it a valuable tool for the future. However, ensuring that developing countries have the necessary capacity and resources to tap into this potential is crucial. Generative AI can truly transform industries and bring about positive socio-economic changes when effectively implemented.

Tomoyuki Naito:
Ladies and gentlemen, good evening. I know this is today’s last session, that’s why not over 100 people coming, but actually I hope you can enjoy, ladies and gentlemen. Welcome to the session, the town hall session, the title, the impact of the rise of generative AI on developing countries, developing economies, opportunities and threats. I’m the organizer and today’s on-site moderator. My name is Tomoyuki Naito, Vice President and Professor at the Graduate School of Information Technology, Kobe Institute of Computing. Very nice to meet you, everyone. Let me just quickly introduce all of the other panels from myself. On my just left-hand side, Ms. Safa Khalid Sariyari, she’s the Senior Business Intelligence Engineer and the Software Engineer, Central Bank of Sudan, the Republic of Sudan, Safa, welcome. And next to Ms. Safa, Mr. Robert Fornon-Cussey, he’s the founding partner and CEO of Orasoft Ltd., the Republic of Rwanda, Mr. Fornon-Cussey, thank you for coming. Actually today we expected to have Ms. Kay McCormack, the Senior Director of Policy, the Digital Impact Alliance, but due to her immediate engagement, she couldn’t make it. So today we have the privilege to have Dr. Sarayu Narajan, she’s the founder of the Apti Institute India, doctor, thanks for coming. And last but not least, on my left-hand side, Mr. Atsushi Yamanaka, Senior Advisor for Transformation Japan International Cooperation Agency Japan, Mr. Yamanaka, thank you for joining us. I’m also looking at Zoom online, we have quite a number of participants online, over 20 people are joining, thank you for coming, thank you for joining today. So as scheduled, and as title, we’d like to begin the title, The Impact of the Rise of Generative AI on Developing Economies, Opportunities and Threats. Let me just begin with a very brief introduction, a very brief explanation of the background of this session. Actually many of you, attendance here and attendance online, as well as the panelists, you have already heard many discussions in the past two days, including today’s discussion throughout the IGF 2023 in Kyoto. Actually the internet for all, or internet we want, internet for everyone, internet for good. This is the kind of the common keyword for today’s, this year’s IGF. On top of that, even yesterday’s official opening ceremony, Prime Minister Kishida of Japanese government, he mentioned about the importance of generative AI, and the importance of the guideline, importance of the knowledge sharing, information sharing, and the collective effort to make AI for good, for everyone. Standing on that kind of background, actually many AI-related sessions have been done, and this discussion, or that kind of the session, is still ongoing until the end of IGF 2023 in Kyoto. Then this session is specifically focusing on generative AI’s impact, specifically for the how it will be impact on the global south. Actually I don’t want to divide north or south, but actually this is very, very important, because there are two aspects, opportunity and threats. On the threat side, many sessions have been discussing about the potential threat on privacy, human rights, or information security. Of course, all those things are very scary. The generative AI correct the information from everywhere, then synthesize all the data into one, as if we don’t know how it is synthesized. But many experts, international experts, as you already heard throughout the session, many international experts already aware about those kind of threats, and they have already shared their knowledge, their worries to everyone, including us, so that the process is already ongoing. I personally have felt that that aspect, threat aspect, is not necessarily we have to fear, we have to worry so much, because many experts or many people have already aware about the other worries, so that we can protect somehow by using the collective wisdom, collective knowledge, collective effort. On the other hand, opportunities. Opportunities-wise, I personally don’t see many discussions are happening in this IGF, but we also know that many opportunities we have seen. So I will invite these knowledge panelists, and I would like to invite all of your opinions, both from on-site and online. Then I would like to allocate first 20 minutes or 25 minutes to hear the opinion from the panelists here, and another half of this session, the last 20 minutes or 25 minutes, I really would like to have your guys’ opinion or comments about opportunities. And of course, threat-wise, you can share your opinion. That would be really, really welcome. So my first question to all the panelists, question number one, is actually go to the other, very fundamental one. Actually, in order for all of you to know about the other experts’ background, knowledge, wisdom, I won’t invite all panelists to answer to us. Question number one. What has been your background in terms of the other related works in ICT sector, in the light of using power of technology to make the better world? So let me just invite one by one, starting from the other, Ms. Fakari.

Safa Khalid Salih Ali:
Thank you, Sensei. I am truly honored to be here today. Thank you, all of us, to attend this session. I am Safa Khalid from Sudan. I start my journey in ICT since 2010, while I’m working in Central Bank of Sudan. My specialization is business intelligence and software engineering. While I’m working in Central Bank, we pass through data analysis for multiple systems that we can see now, generative AI can help us to reduce a lot of time of working, and it help us to make many prediction model for crisis, sometimes happen in any country. And also, when you’re using generative AI, it is a benefit for all Central Bank, which can impact economy of the country, like maintaining financial stability of country, or promoting economic development of the country, in Sudan or in another country. Also now, when you’re thinking about FinTech technology, generative AI can help us in customer experience and can build us many customer experience system, which can address an emergent situation for customer, instead of waiting for next day to go to the bank and find solution for your actions, and also enhance any part of analysis tool and prediction for any module. This really happen, for example, in the last year, before the world start in Sudan, when we have auction, you can use generative AI to build simulation for the policy, instead of just embracing and using the policy directly in the economy. This simulation, if you implement it, it can help us to find the weaknesses of the policy, instead of using directly to your economy. Thank you.

Tomoyuki Naito:
All right, thanks very much. Okay, Mr. Robert Ford, please go ahead.

Robert Ford Nkusi:
Thank you. My name is Robert Ford. I’m currently leading a software testing qualifications team in Rwanda, under the Rwanda Software Testing Qualifications Board, one of the members of the International Software Testing Qualifications Board. But before that, I was supporting the government of Rwanda under the United Nations in designing the implementation plan for the Child Online Protection Policy, which Rwanda designed a couple of years ago. And it’s now a framework that is helping the country in setting proper guidelines for safety for child online engagement. Before that, I participated a lot in the regional framework for the one network area for the East African region. For some of you who may not know the EAC framework, the East African community is composed of six countries, Uganda, Kenya, Rwanda, Tanzania, South Sudan, and Burundi. And at some point in the framework, they wanted to create one network area, having inbound, inbound traffic, data traffic, as one network traffic, so that they can circumvent the costs of data roaming for people in the region. So in a couple of years, we were able to set the one network area, the first actually proof of concept ever tested around the world. And in that, I was specifically more on the component of cross-border mobile financial services, helping citizens in the countries in that region to transfer and transact financially between nations using what is popularly known as mobile money. I have participated at AFRINIC, the Internet Numbers Registry for Africa. But after that, I’m now currently engaged with JICA, the Japan government, in implementing a project in Uganda. It’s an ICT industry promotion project. It’s a four-year project with four major outputs that are supposed to help the country build capacity, strong capacity in the ICT industry.

Tomoyuki Naito:
All right, thanks very much, Mr. Ford. Okay, Dr. Salayu Natarajan, could you go ahead, please?

Sarayu Natarajan:
Thank you very much. Thank you for enabling me to be a part of this conversation. It’s a very important one, both in understanding the disadvantages and advantages and some of the risks of generative AI. Thanks also to the audience. I know it’s 6 p.m. here, and it must be a range of different times across the world, so thank you for joining online as well. My name is Salayu Natarajan. I am the co-founder of Apti Institute, which is an institution that works on questions at the intersection of technology and society. We have three big areas of work. We focus on algorithmic and platform-mediated work. We look quite extensively at digitization and digital infrastructure and the ways in which governments and states can reach their citizens, particularly on questions of sustainability financing, governance, replication of digital systems across the world, and also extensively on questions of data and data governance. And AI, and particularly generative AI as a theme, is one that cuts across all of these areas, and we’ve been exploring it quite significantly, hopefully more over the course of this conversation. Back to you, Dr. Naito.

Tomoyuki Naito:
Thanks very much, Dr. Harari, and Mr. Atsushi Yamanaka, please go ahead.

Atsushi Yamanaka:
Thank you so much. Well, it’s very hard to actually be so very concise introductions to my predecessor now. Thank you so much, actually, for joining this session. You guys actually are very brave, because you actually could not actually resist the urge of actually having a GIZ reception downstairs. So thank you so much, actually, for your effort of coming to this session. My name is Atsushi Yamanaka. I’m actually a senior advisor on digital transformations at JICA. At JICA, we’re actually trying to promote the X for the improved well-being of all right now, incorporating a lot of actually these technology elements into different projects and then support initiatives. But prior to that, I’ve been actually doing this field for quite a long time, actually. This is my 28th year. In fact, it’s like I feel so old about that, of pursuing ICT for development. Initially, like I was in UNDP, and then I actually was involved quite intimately in the WSIS process. So it’s really personal for me to be here, and it’s really happy to see IGF actually came to Kyoto, and also discussing about how this process is going to actually help, finally, hopefully, change the world for a better place using these technologies. And AI certainly is one area where it’s going to have a lot of potentials, and also a lot of threats as well. So this conversation is very timely, and I’m really happy to be part of this. Thank you.

Tomoyuki Naito:
Okay. Thanks very much. So for the sake of time, let me just quickly go to the other core of today’s session. This is the core question to all the panelists. My second question to all the panelists. Do you think that the rise of generative AI represented by chat GPT is a good thing to the economic and social development aspect of developing economies? Please answer by yes or no, and with your very succinct reasons, please. Maybe let me just start with Yamanaka-san.

Atsushi Yamanaka:
Thank you, Naito-sensei. Can I say maybe? Or depend? Well, I understand that you want a very precise answer with yes or no, but I don’t think I’m qualified to say yes or no. So would it be okay? Yeah, it’s okay. It’s up to you. It’s okay. Thank you so much. Yes. Well, in a way, yes, because we actually had a colleague, actually. We sent him to Japan. He studied AI. He actually did a PhD here in Japan, and he was actually doing research on generative AI models for African language. Now, he was actually in RIKEN, one of the top research institutes in Japan, but now he, unfortunately for Japan, he actually moved to Princeton to continue his work. But having this kind of local language model actually incorporated into the generative AI could really open up the opportunities for those people who are actually unconnected or not digitally involved in it. Because of high barriers in terms of digital literacy, this inclusion has been a major challenge. And then this has been the issue for the last 20 years. And the last 2.6 billion people is going to be the most and hardest people to reach. So for that, I think AI, and specifically having the local language generative AI models, could have opportunity to open up the opportunity for them. So that’s yes. The no part, yes. Of course, there’s a threat. We talked a lot about, in terms of misinformation, malinformation, disinformation. There is even industry in, I think it’s northern Macedonia, where this village actually churned out all these malinformations. It’s a business. So people who want to actually have this malinformation, they actually hire these people in this particular village and it became an industry. And it’s getting very, very difficult for us to distinguish if any information is actually real or not. So that, I think, is going to be a huge threat in terms of information accuracies and in trust that we actually have in the internet or to the digital technology as a whole. So that’s actually yes and no. And I’m sure that my fellow participants also have a similar yes or no moment.

Sarayu Natarajan:
Thank you. I will take, again, a response which is yes if we attend to questions of governance and regulation. I do think, I mean, yes and no, yes if we take a, I mean, I feel like to think about generative AI without thinking about the consequences and the harms, which occur at all different levels, both in, or injustices of generative AI occur at several levels. One, there is the level of labor. Generative AI does not exist without the labor of several workers in very many parts of the developing world. So to talk about it abstracted from the way in which generative AI is itself created, which is the labor of data annotators and labelers, that might be a bit limiting. And so that needs to be a part of the conversation. Second is, I think, the way in which data itself is structured, which is that it often limits the presence of data pertaining to gender, race, et cetera, may limit the capability it is generative AI, so the applicability and use in context in the developing world may be limited. I think the third, of course, is what you mentioned, Mr. Yamanaka, which is around the consequences such as disinformation and misinformation, which generative AI makes very easy. So with this framework of the kinds of injustice that generative AI may bring about, we can start to think about both what are the use cases and how may we govern them to ensure that all of us have meaningful digital lives and digital futures. I’ll pause here and look forward to the rest of the discussion.

Tomoyuki Naito:
Yeah, thanks very much. That’s a very good point and a very important point. Mr. Yamanaka emphasized about the local language issues and Dr. Sara, you mentioned about start with the use case, then we can deepen the more discussion. That’s very important and significant case. Then, Mr. Robert Ford, your opinion is always very important. Thank you.

Robert Ford Nkusi:
How can I possibly, can I think I have a good voice? Thank you. Thank you, VP. Just a day before I flew here, my niece asked me and said, so when I showed her the topic I was going to discuss, I was going to be a panelist at this conference, she asked me a very intriguing question that I kept asking myself over and over when I was flying. She asked me, said, uncle, don’t you think the digital technology that we are consuming today has come earlier than it should have come? That is the world capable and the people capable to consume and use the technologies that we have today profitably for their own benefit. And I kept juggling my brain back and forth to see if that was right or wrong. So, now talking about generative AI, it’s better to tackle this topic when we have deeply digested and understood what this animal is. When I took my class many years ago on computer science, we always thought, when is it going to be a time when technology will take away from us the responsibility to write lines of code so that something else does it? And AI now is with us today. So, to answer your question, is it good or bad? Are we looking at dangers or are we looking at a comfortable world? In the time ahead. Like Sensei, yes and no. One, I’ll just speak specifically one example. The nature of predictability of the power of AI is going to help us, especially the global south, in being able to determine the kind of content that goes online before we can even know the danger that content is going to produce to us. And I speak this from the line of authority that comes from the Child Online Protection, for example. Today, when we look at how we struggle and fight to keep our children safe online, before we put mitigation measures, the content is online and children are consuming this content. Generative AI now equips us with the capability to mitigate that. That we can be able to use those tools to mitigate that. That’s the positive part. But the negative part, some of the tools, like ChatGPT, gives you so good response that many times we don’t even need to question it. That each time you ask, the response is so accurate to your understanding that you don’t even want to question that. But behind that text, there is a lot that can be questionable, which now puts us at the crossroads of what is good for us, what’s not good for us. Because the way generative AI works is that it’s just using machine learning to train some software and data to be able to give you what it gives you. That’s how it works. So if what we get seems as too good to be nice, then we don’t question. And then we stand at the edge or at the rift to fall off into oblivion by technology. That’s just the tool. Thank you.

Tomoyuki Naito:
All right, thanks very much, Mr. Ford. Hey, Safa, please go ahead. Your opinion.

Safa Khalid Salih Ali:
Okay, let me answer by yes. Because I am ICT for all this long, I will directly say yes. It can help you. When you’re thinking for economic and you’re thinking for developing country, how long will it take to analyze just the data set? It take a lot to get insight from data set. But by using generative AI, we can go to insight directly and get impact directly to economy instead of wasting a lot of time in routine work of just analyze this data. When you’re thinking about that, thinking about how can we save in the cost, how many employee you needed to just analyze this data. By generative AI, we can found your insight directly and it’s helpful. Also, when we thinking about economy, you need to think about financial inclusion and FinTech. To deserve real financial inclusion by this way of traditional way, you can deserve it for all the citizen of any country. But if you have like generative AI, we can have opportunity to all participants to find the service you need like using virtualization access, like using chatbot, any type of AI tools. And when we thinking about that GBT and which can allow you to analyze the data and thinking about how many time you need to just make risk assessment for any bank, for any customer you have it in the bank to give him just loan. By using generative AI, we can make fraud detection and you can make risk assessment and build model which can predict for you what’s the habit of this customer based on the historical information. For all these reason, I can say for your question, yes, directly. Yes, we have drawbacks, but it can cover it by any guidelines. Thank you, Cesar.

Tomoyuki Naito:
Yeah, thanks very much, Ms. Alfa. Actually, to the audience, actually as you fully understand that we have a variety of the nature of the countries as well as the nature of the jobs. Every partners have different occupations. Then it is quite interesting and quite significant and it’s quite important. Looking at the power of AI or looking at the threats of AI from different angles, it is quite important discussion, I personally believe. Then as a result, four partners here mentioned somehow yes side. No one say obviously no. So if I understand like that, four partners over here say yes somehow. Let me just ask one more question. While G7 countries, G7 member countries including Japan are currently preparing the basic guideline of appropriate AI use in societies, do you think that developing and emerging economies should also prepare original guidelines for the AI use? How do you think about that? Let me just ask this question to all the panelists. Let me invite Mr. Robert Ford first. Robert?

Robert Ford Nkusi:
Okay, I went to the computer science class but not to the audio class. So thank you. Each time we have a discussion that draws a line between developing and developed economies, I struggle. I struggle to maintain that classification. And I will quickly give an example that the one part of the world is struggling to understand or to draw lessons or to draw benchmarks from some of the success stories from another class of people we are talking about. So when we talk about the developed world, and I’ll give just a simple example, the developing world, what they call developing economies. For very many years, many years, for as long as I can remember, the world’s financial systems in the developed West were based on data that runs on plastic cards, right? That each one of us is carrying a wallet with so many plastic cards. For Africa, in the, not more than a decade ago, we leapfrogged from that. And for, they are able to transact using mobile money. And they are circumventing the whole trouble of the environmental impact of keeping plastics in people’s wallets. If we were to collectively remove plastics from people’s wallets, and we pile them together, we probably could fill up a country. And this is a very bad effect to the environment that we live in. But the other part of the world is failing or is dragging its feet in quickly benchmarking on that and say, why don’t we have a plastic-free world, at least from our simple cards? Then we can be able to transact with mobile money. So just giving examples. So when the G7 is discussing about legal and regulatory framework, about generative AI, they probably need to go down there and see what is it that there is down there that they can pick from. I know countries in Africa that have moved far away in designing and crafting regulatory frameworks for AI and for these other technologies. And they are very fast in moving towards that. So at some point, I don’t think it requires to be a member of the G7 to determine the kind of policies that are going to guide and mainstream thought process towards how we consume technology for a safer world tomorrow.

Tomoyuki Naito:
All right, thanks very much, Mr. Ford. Can I invite Dr. Sayoni? Can I invite Dr. Sarayu? Yes. Please.

Sarayu Natarajan:
Thank you very much and thank you for that. I largely agree. I think that not just because of the several advances made in several parts of the world in terms of thinking AI, I do not think this is a conversation that is entirely unique or the problems are entirely unique to parts of the developing world. I think one of the strange things about technology is that it’s a great equalizer, both in positive senses and some of the more harmful ones. And I think for us to think about these as unique problems from a frame of exceptionalism may be limiting. And so building out frameworks of governance, frameworks of regulation from an inclusive standpoint, which includes several global participants is very necessary. My yes if, to go back to that was deeply qualified, but I will speak to one specific theme that I mentioned in terms of the injustices of generative AI which is the question of labor. And I think there are two sides to the question of labor in the context of generative AI. The first is that labor and human labor is very critical to building generative AI. So if somebody doesn’t label the data that is used to build a large language model, a large language model does not exist. And this is true of image models as well. And much of this labor is in very many parts of what is called as the developing world. Now, those who build AI in that sense that are responsible for the labeling, annotation, marking, categorization of data are never really a part of the conversation. So that’s one significant injustice. So while there might be frameworks for governing regulating generative AI use, the way in which generative AI is made itself has a significant geopolitical component. The second dimension of generative AI is a much more downstream effect which is the question of job loss. And to the point Mustafa made, job loss is a complicated question that has different meanings in different countries. In mine, which is India, we have a population of 1.4 billion where challenges around thinking about employment, job losses are very significant. India is also an IT services company, so a large part of the economy relies on IT service provision. And again, the question of job losses through the use of services like charge GPT and generative AI more broadly speaking is a very significant one. All of this is to say that conversations about generative AI without starting from the how are we going to govern some of the harms just as much as how are we going to deploy it might be limiting. So we have to think very carefully about use cases, very carefully about first order, second order, third order consequences of the use of generative AI. That is not to say that there is no use case at all. There are several applications for it. But governance is a part of deployment is where I’m coming from. I’ll pause here and.

Tomoyuki Naito:
All right, thanks so much, Dr. Actually, thanks very much for you touch upon the other aspect of job itself. Actually, just to share the information to all the other participants here on site and online. Recently, ILO, the International Labor Organization just released almost one month plus ago, sometime late August. They just used their ILO model to analyze the impact of the generative AI adaptation to the 59 countries as a model. Then, according to their result, the impact of job loss aspect is much heavier to the advanced economy than much less in the developing economies. That means a lot of meanings are contained in that analysis. But I just strongly recommend if you are interested in their job analysis or the job impact of the generative AI, ILO have released a very good working report back in August. But anyhow, if I come back to the question, let me just invite Yamanaka-san for your opinion.

Atsushi Yamanaka:
Thank you so much. Actually, there’s a lot to think about, yes. I agree with the doctors and I agree with Robert about, yes, this is actually, we don’t necessarily want another fragmentation, right? We had a discussion yesterday about data flow or data governance, and we had the sessions, and this question also came up. Do we actually need to create another model or another regulations specifically for emerging economy? Answer probably no. Why should it be? Instead, I think the so-called emerging nations should be part of the global discussions on the framework or like case makings instead of actually trying to fragment. It’s already a fragmented world. Internet has been fragmented. There is also like agenda is fragmented. Why do we actually need to further fragment this field? So AI regulations, whether we can succeed or not, regulations, I’m not sure if we can do this, because there’s so many different opinions, but I think it is really important to have this multi-stakeholder approach in terms of actually having the opportunities for the emerging countries to actually have their voices and inputs into the formulations of a regime or mechanisms to regulate or to use AI for better word. I think that’s, I think, is one of the things we need to do. Going back to, I think, it was labor side. I think that’s gonna be a very, very important component. That’s something that I should have said in terms of potential negative aspect of generative AI. Already, like Robert was mentioning about software engineers, or even like the illustrators, or even like storytellers, because I think there was a huge actually strike actually doing the Guild of Screenwriters Guild in the United States, because they were saying, okay, generative AI can do their jobs. Now, I think we’re gonna have the Luddite movement. Do you know what the Luddite movement? That actually happened in the beginning of 19th century when the industrialization actually came into UK. All the workers, they start actually destroying the machines. I think we’re gonna see that if we do not actually come up with a model which is conducive. Another thing is if we do not now give them the opportunity for them to change their business models or their skills, re-skilling them, if you don’t do that, I think we’re gonna see another Luddite movement for the AI.

Tomoyuki Naito:
All right, thanks very much. Re-skilling, another word which the government of Japan is really emphasizing domestically. Okay, Safa, your opinion about my question.

Safa Khalid Salih Ali:
So we need to think about what we need to do to make sure that we have a system that can be used for the development of AI. According to guideline, when we’re thinking about guideline for developing country, we need to think about what we already said about data challenge. Really, we face very critical issue in data in developing country. We can’t find system with fully clear data. So we need to think about what we need to do to make sure that we have a system that can be used for the development of AI in developing country. Maybe like this issue we didn’t found in G7 country. For that, you need to put it as a challenges for these guidelines. Also, we need to think about the socioeconomics between this country and G7 when you compare. It is not about the fragmentation of the guidelines, but guidelines, you need to put in mind also the capacity of building. You need to think about how can you solve the issue of infrastructure. You can’t put just generative AI solution for someone who can’t use it. Also, we need to use to think about the capacity building. A lot of people in developing country didn’t know about anything about generative AI. You can’t use it. Maybe it’s a big company, but they didn’t know about it. So, we need to think about how can we use generative AI in developing country and how can we use it for the development of the country. The third issue is data privacy and labor, as you said. Do you think wasting time to make routine jobs, which you already can do it in minutes, instead of thinking about the productivity and quality of the work? If you have stuff and can focus more on the quality of the work, it’s better to use generative AI in the routine work. So, we need to think about how can we use generative AI in developing country and how can we use it for the development of the country. For example, if you are thinking about central bank, you think about if crisis happen anywhere, you need to calculate what’s the impact of this crisis to your own economy. With generative AI, you can use this solution to be prediction, to make any prediction for what the next action will be. So, we need to think about what’s the next action, what’s the next action, because it can be emergency action, you need to take it instead to go in the drawbacks directly.

Tomoyuki Naito:
Thank you. ≫ Thanks very much, a very important aspect. You beautifully mentioned about the opportunity side. Now, I would like to open the floor for the question, and our panelists to ask their questions. So, please, gentlemen, come to the microphone. I already have the question online, but let me just prioritize on-site question first. Please, kindly say your name and please the question.

Audience:
≫ So, my name is from Deutsche Welle Academy, Germany. I have so many questions, I don’t know which to ask first, because I recently heard a professor from Ghana. He said, you know, we have a kind of democratic system. We have a kind of democratic system, but we have a kind of neutral political system. So, I wonder why you are not mentioning this topic. But my question is, so, we are discussing, like, as we live in a neutral political system, like, we have a kind of democratic system all over the world. So, I’m wondering, like, what do you think about this kind of democratic system, like, in Europe, also, and non-democratic states worldwide, especially also Sudan, South Sudan, and your region. So, my question is, what do you think, will these technology, especially in the age of AI, will lead us to more freedom and democracy, or will it be the opposite? Because if the authoritarian rulers want to be free, they will be free, but if the authoritarian rulers want to be free, they will lead to more free societies. So, this is my maybe too big question, but… ≫ Yeah, thanks very much.

Tomoyuki Naito:
Big question, but very important question. Any panelists want to answer to this question?

Robert Ford Nkusi:
≫ Thank you. I think it depends on how you define those values of democracy, and authoritarianism, and whatever you want to call them. In my country, if, say, a country, for example, if you have a country, let’s say, in the United States, if you have a country, let’s say, in the United States, if you have a democracy in India, or in the United States, what does a democracy mean? For example, if, say, a President stays in power four, five, six times and is doing the right thing, to me, that’s okay. So, it’s Germany, for example. Germany does not have time limits, right? Germany can have a head of the country, and, in fact, if you look at the United States, if you look at the United States, if you look at the League, if you look at my country, it’s different from how my country is going to look at Germany. So, in Uganda, or in Rwanda, or in Kenya, if the British system says you can stay in power for as long as your party is keeping you in power, that’s okay. It’s not authoritarianism. It’s not authoritarianism. It’s not authoritarianism. So, and there should be a way in which we validate political leadership. At what point do we define this as being authoritarian, or not working in the interest of the masses, and I’m not saying it’s not fair. I agree with you. We have countries that are not working in the interest of the masses, and we have countries that are working in the interest of the masses, and, in fact, we have countries that are not working in the interest of the masses, meaning, the power of predictability that generative AI gives us is so immense that with it, we could build great potential that can change the way we run politics today and in the future. But, for authoritarianism, and dictatorship, and God knows what, I have good friends who have been dictatorial for 30 years and I never encountered any example in more dictatorian country because they will keep someone powerful as long as God knows what. In Britain. I know countries in Africa who have very good elections, of course.

Atsushi Yamanaka:
Not good, thank you. I think it’s a good thing. I think it’s a good thing. It’s been like with IGF, someone was calling, this IGF is so banner, there’s nothing controversial. No one is making anything controversial. Thank you so much for asking those kind of questions. It’s interesting questions. For me, I also came from this party where actually we have so-called free election system, but I was told that it’s not true. It’s not true, because I actually asked questions about China, because Chinese actually have the systems of scoring, right, Dr. Song, yes? Now, I was, you know, for me, it’s a bit difficult to accept that kind of system where I will be watched and scored every single moment. I was told that, you know, the Chinese actually have the system of scoring, but they don’t have the money to bribe the Chinese, you know, because they do not bribe people, they don’t have money to bribe or they don’t have the connections. For them, they said, well, now, if I actually are very good citizens, if I actually, you know, do something good for the society, my score goes up, and I actually get different opportunity which I never actually got. So, you know, I don’t have the answers to it. But at the same time, I feel like if, you know, we can keep the privacy, you know, if we can have, like, basic human rights, okay, last ten minutes, so we have to be quickly. I think, you know, how the polity would actually, you know, manage, you know, not control, but manage the society, I think it should be depending on the society itself, you know, we are here, you know, because so-called developed countries, we are not here to judge them, that I feel very strongly, you know, working in developing countries for so long. I don’t know. I may get hammered by this, but ‑‑

Tomoyuki Naito:
. ≫ Thanks very much. Yeah, thanks very much for the very good interaction, which going beyond my expectation, but anyhow, thanks very much for the very good question and answers. Then I have a question. I have a question about, you know, how can we apply generative AI in developing countries? I have recognized several, actually, people raise hand online, on Zoom. Then before that, let me just check chat box, and I got the one question from the Mr. Mohamed Hanif Garanai, he is questioning about could we apply generative AI in developing countries, like I do in the audience. I think perhaps we can fill the opinions online. Some of the online participants kindly answered already, so thanks very much for the online interaction spontaneously. So let me just actually ask a question. I’m not sure if you can hear me. Can you hear me? Yes, I can hear you. Please go ahead. Thank you.

Audience:
Now they’ll let my video come, so I’ll join all of you in person. Hello, I’m Debra Allen-Rogers, I’m in the Hague. I’m originally from New York City. I have a nonprofit here called Find Out Why. It’s a digital fluency lab that promotes digital fluency. I was asked to ask a question and even the sort of spicy comments that followed about Germany being dictatorial, which I don’t think is obviously true in the context of this. But I did want to ask the question about if we could rename, I know this is going to sound naïve, but just look at my age and know my background in design and so on. I’m going to ask this question. Developing countries, we’re in a transition now, and we can do that in highly functional societies, because the highly functional digital societies, for example, let’s say in Estonia, in Finland, in Norway, I’m living in the Netherlands, highly functional digital societies are great places to come learn for young people on travel expeditions. In Japan, back at post-war, there was that, I was looking for the name of the program, but sent students and young people around the world to develop best practices, to observe and to learn and then bring it back to Japan and design the society how they wanted it to be after the war. So aren’t we at another juncture like that right now, that India has a lot to teach, the continent has a lot to teach, MENA has a lot to teach, and those societies have young people, 70% are under the age of 30 going forward. So we have to redefine, we ourselves working in this right now, have to redefine some of these terms so that we don’t fall back on developing world countries. It’s a new day, and we don’t have to spend so much time talking, I’m kind of breaking my own rule by talking about it so much right now, but I’m saying that we should redefine it when we give our speeches and when we talk, and also part of the work I do is to help young people travel the world to see best practices, bring it back home, and then decide how you want to design your own society, but we do have highly functional digital societies, Estonia, Taiwan, the ones I said in the Scandinavian, the Nordic countries, that are right there for us to look at and watch. Thanks very much for taking my question.

Tomoyuki Naito:
≫ Yeah, thanks very much, Ms. Rogers, great comment. And one more person or organization, Ghana IGF remote hub, you are raising hand. Please go ahead, if you can hear my voice, please go ahead.

Audience:
≫ Okay. I’m Dennis. I’m from Ghana. My question is, with the use of native AI to create, sorry, does the native AI do good or affect human creativity? And if yes, how and what contribution is it making to the human resource? And the second question is, with the first group of native AI, help or not, how do you think about the use of native AI in developing countries with fewer resources and technology? Could it boost their development or might it make the technology gap between countries even bigger? Thank you.

Tomoyuki Naito:
≫ Very sharp question. Any panelists who want to answer? ≫ Hello. ≫ Doctor? ≫ My name is Joseph. I’m from Ghana. I’m from Ghana. I’m here to talk about the use of native AI in developing countries with fewer resources and technology. Can I just invite the doctor to answer the first question?

Sarayu Natarajan:
≫ I’ll try answering very briefly to the first question on creativity. I think the second one around development gaps has been tackled a little bit, but with respect to human creativity, it’s hard to talk about the effects on creativity, but I think it’s important to think about creativity as something that’s self-built on the back of creative work by artists, by sort of, you know, producers and generators of music, makers of music, and to think about creativity as abstracted from some of how generative AI is made may be limiting. But it could help in the sense that it could be a new way of thinking about creating, you know, certain types of writing literature based on which further human creativity and ingenuity is applied, but it could also have some deleterious effects in the absence of efforts to for the It should not supplant in my mind the efforts of the education system of, you know, human society at the time, and skills such as those in young minds. So we should not see this to my mind as a zero-sum game. And particularly the question of creativity must look at how it was born. So thank you.

Tomoyuki Naito:
≫ Thanks very much, Dr. Sarayu. You want to answer to the second question?

Atsushi Yamanaka:
≫ Summarize the questions on the second one. Basically trying to actually, if the AI technology is going to be useful for the society, or the gap, gap, gap, between those. That’s a good question. I think earlier Dr. Sarayu was mentioning about technology being, you know, especially the new technologies actually reducing the gap. Because we could actually, you know, there is no sort of hindrance or the interaction sort of barrier is much lower in the digital technologies than many other technologies. So I think, you know, I don’t think that’s going to be a problem, you know, because we don’t have a lot of technology, you know, so I think it does not necessarily be that, you know, developing countries are going to lag behind. And rather, I think it’s going to be more interesting. A lot of innovation is actually coming from so-called emerging economies right now. And we may actually see much more innovation coming from the emerging economies rather than these things coming from the so-called western countries. So in that respect, I don’t necessarily think the developing country is going to be lagging behind, but rather I think with new digital technologies, I think there’s more contributions. I don’t necessarily like the word reverse innovations because that’s very pretentious, but I think the innovations coming from the emerging economies, developing countries, I think they are going to be the trend that we see in the future.

Tomoyuki Naito:
≫ All right, thanks very much. Actually, we have only one minute left up to the end of the session. So apology for the other participants online for giving me the other question. Let me just summarize today’s session very briefly. As Mr. Yamanaka’s last comment mentioned, actually, we don’t have to look at only the threat side. Actually, the opportunity, opportunity which every country can utilize the new technology as a sort of innovation to leverage more the economic development as well as the social development is the key to discuss more continuously, and is the key to emphasize not only G7 countries, the leading, the guidance, and so on. But also, the other countries, the leading countries, the leading countries, the leading countries, for instance, so other, you know, more than 190 countries can do proactively to utilize it, to do it for your own business, your own countries in the future. So that could be the main message of today’s session, and I’m sorry about my poor time today, but I think it’s time to wrap up the session, so thank you very much. Thank you very much, everyone, for being here. Let me just conclude this session, since time is already up. So please join me to give the round of applause to all the panelists here. Thank you very much. And thank you, all the other participants, on-site and online. Thank you so much.